Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: brant test
From
Richard Williams <[email protected]>
To
"[email protected]" <[email protected]>, "[email protected]" <[email protected]>
Subject
Re: st: brant test
Date
Wed, 10 Mar 2010 08:29:15 -0500
You could also try the -omodel- command available from SSC. But my
guess is you need to simplify the model. Maybe cut some variables or
combine some categories. This sort of thing happens when the data
are being spread too thin. The gologit2 troubleshooting page is at
http://www.nd.edu/~rwilliam/gologit2/tsfaq.html . See especially the
section entitled "gologit2 is very slow and/or does not
converge/and/or produces implausible estimates."
At 03:37 AM 3/10/2010, Katrin Gasior wrote:
Hi,
after carrying out an ologit, I wanted to test the parallel regression
assumption by using a brant test. The problem is that the brant test doesn't
work (error message: not all independent variables can be retained in all
binary logits brant test cannot be computed)
Thus I wanted to solve the problem by using gologit2 and autofit as an
alternative. BUT this command takes forever. Even after waiting for about 5
hours, I still don't get a result.
Does anyone by any chance know how to solve the problem. Or knows an
alternative how to find out which independent variables are violating the
parallel regression assumption?
Thank's for your help, best wishes
Katrin
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/